Snap Inc.

Samsung Electronics Co., Ltd.

Dolby Laboratories Licensing Corporation

Adeia Guides Inc.

Adeia Imaging LLC

Kwai Inc.

Siemens Medical Solutions Usa, Inc.

Siemens Corporate Research, Inc.

Ning Xu

Jia Li

Jianchao Yang

Yeong-Taeg Kim

Tao Chen

Rahul Bhupendra Sheth

Kevin Dechau Tang

Yuncheng Li

Yuke Zhu

James E Crenshaw

Samir N Hulyalkar

Serhad Doken

Sumant Milind Hanumante

Yanen Li

Timo Kunkel

Jean-Yves Couleaud

Zhou Ren

Leonardo Ribas Machado Das Neves

Enxu Yan

Nathan Litke

Daniel Szeto

Tan Yu

Jose Jesus (Joseph) Paredes

Reda Harb

Scott J Daly

Linjie Luo

Guohui Wang

Bongsun Lee

Kirk Ouimet

Shenlong Wang

Dhritiman Sagar

Dhananjay Lal

Ning Zhang

Sangkeun Lee

Xing Mei

Kun Duan

Chongyang Ma

Jaewook Chung

William Brendel

Vitor R Carvalho

Luis Carlos Dos Santos Marujo

Xiaoyu Wang

Xin Chen

Daniel Ron

Yu Hu

Venkata Satya Pradeep Karuturi

Jinhua Jiang

Shibi He

Yingzhen Yang

Jinxi Guo

Jianfei Yu

Alfred She

Jian Ren

Anup Basu

Limin Liu

Lawrence Jason Muhlstein

Haiying Guan

Robin Atkins

Gregory John Ward

Kimball Darr Thurston, Iii

Zhi Zhou

Kevin J Stec

Zhiyun Li

Thomas J O'Donnell

Paras Maharjan

Gopal Erinjippurath

Richard D White

Georgios Georgiadis

Raymond Yeung

Randolph M Setser

Suzanne Farrell

Aldis Sipolins

Ravi Bansal

Toshiro Ozawa

Yuyan Song

Fei Wu

Jinyu Yang

Jingjing Liu

Xuan Xu

Arunkumar Mohananchettiar

Jia Lia

Kimball Darr Thurston Iii

Systems and methods relate to improving image quality of panoramic images. A system generates for display on a device, a user interface for capturing a panoramic image, receives image data from a camera of the device from a plurality of view angles, and generates for display on a device a preliminary panoramic image based on the image data. The system determines that a position corresponding to a particular view angle of the preliminary panoramic image is problematic, and in response to the determining, generates for display an indicator for the preliminary panoramic image prompting for the camera of the device to be repositioned to the particular view angle of the preliminary panoramic image that was determined to be problematic.

Methods and systems to improve image quality and automate panorama shots

Systems, methods, and apparatuses are described herein for obtaining an image, wherein the image includes a digital watermark that embeds metadata associated with at least one object in the image, and wherein the metadata comprises an indication of a location of the at least one object in the image. The embedded metadata may be extracted based on the digital watermark. Based on the extracted metadata, a portion of the image corresponding to the location of the at least one object in the image may be caused to be selectable to access supplemental information related to the at least one object.

Systems and methods for extending selectable object capability to a captured image

Systems, devices, methods, media, and instructions for automated image processing and content curation are described. In one embodiment a server computer system communicates at least a portion of a first content collection to a first client device, and receives a first selection communication in response, the first selection communication identifying a first piece of content of the first plurality of pieces of content. The server analyzes analyzing the first piece of content to identify a set of context values for the first piece of content, and accesses accessing a second content collection comprising pieces of content sharing at least a portion of the set of context values of the first piece of content. In various embodiments, different content values, image processing operations, and content selection operations are used to curate the content collections.

Content navigation with automated curation

System and method are provided for sharing a counterpart version of a media asset. A media application generates for display a first media asset on a first device associated with a first user profile. The media application determines that a second device associated with a second user profile is generating for display a second media asset and the second user profile is designated to share a counterpart version of media consumed in association with the second user profile with the first device. The media application identifies a time to share a counterpart version of the second media asset. At the identified time, the media application generates for display an overlay over the display of the first media asset on the first device. The overlay comprises the counterpart version of the second media asset and a plurality of user interface options.

Systems and methods for sharing media items

Systems and methods are provided herein for manipulating a stereoscopic display so that portions of the stereoscopic display are located at varying planes. This may be accomplished by a device comprising a plurality of microdisplays receiving a piece of three-dimensional (3D) content. The device may determine that a portion of the 3D content comprises a 3D object at a location within the 3D content. The device may then change the position of a first microdisplay according to the location of the 3D object within the 3D content. The device may display the 3D content using the plurality of microdisplays, wherein the first microdisplay displays the 3D object. The first microdisplay changing positions according to the location of the 3D object within the 3D content provides depth for rendering the 3D object at a suitable accommodation distance and reduces or eliminates the vergence-accommodation conflict for the user.

Adaptive displays to address the vergence-accommodation conflict

Systems and methods are provided for a camera and use of a camera to capture image data from 3D scenes and thereby generate large depth of field visualizations. The camera includes an optical system configured to generate an image of the 3D scene, an image sensor array positioned to receive the image of the 3D scene and including a plurality of image sensor segments, and a plurality of extensible members, each extensible member coupled between one of the image sensor segments and a housing and configured to move each coupled image sensor segment parallel to an optical axis of the optical system. In the process, the image data generated by the camera is captured and analyzed by control circuitry, and further movements of the image sensor segments are based on the analyzed image data.

Large depth of field camera with adjustable image sensor array

Systems and methods are described for replacing an object being presented in a livestream with a secondary object that is personalized to a user that is consuming the livestream and rendering the livestream with the secondary object are described. The methods identify a target object in a livestream based on certain selection factors. A secondary object that is contextually related to the target object and selected based on selection factors is selected. A 2D-to-3D conversion of the scene description is performed to generate a 3D model. A replacement option is selected. The attributes of the secondary object based on its 3D model are matched with attributes of the target object based on its generated 3D model. Quality checks and scalability options are explored. The livestream is re-rendered with the secondary object having replaced the target object.

Systems and methods for switching between alternate scene objects during livestreaming

Systems and methods are described for enabling a lensless camera having an image sensor and a mask to be positioned behind a display screen of a device, which allows for the device to have an increased screen-to-body ratio. The image sensor captures an image based on the light that travels through the display screen and the mask. The display screen may include portions between pixel elements that allow light to pass through. The mask may include a pattern, such as an opaque material with portions that allow light to pass through from the portions of the display layer to the image sensor. The image captured by the image sensor may be indiscernible to humans. The system may utilize a trained machine learning model to reconstruct the image, using data about the pattern of the mask, so humans may visually recognize features in the image.

Lensless camera positioned behind display layer

Methods and systems are disclosed herein for modifying one or more color changing exterior surfaces of a vehicle in response to identified safety events. Specifically, a safety event is identified in a surrounding environment of a vehicle that comprises one or more color-changing exterior surfaces. In response to the identifying of the safety event, a color scheme is selected for the one or more color-changing exterior surfaces of the vehicle based on the type or safety event and the appearance of the surrounding environment. One or more color changing exterior surfaces of the vehicle are modified to match the selected color scheme.

System and methods to modify active car paint

Methods and systems are described for generating memes and enhanced content in electronic communication. The communication is received, analyzed, and assessed for determining whether the enhanced content should be included in the communication. A content, context, intent, and/or mood of the communication is determined. For meme generation implementations, images and captions are generated based on the communication. Some candidate memes incorporate an image of one of the participants in the communication. Enhanced content candidates are generated and presented to the user for inclusion in the communication. Artificial intelligence systems, including neural networks, and models are utilized to improve the enhanced content generation. Related apparatuses, devices, techniques, and articles are also described.

Generating memes and enhanced content in electronic communication

A computer-implement method comprises: training a classifier with labeled data from a dataset; classifying, by the trained classifier, unlabeled data from the dataset; providing, by the classifier to a policy gradient, a reward signal for each data/query pair; transferring, by the classifier to a ranker, learning; training, by the policy gradient, the ranker; ranking data from the dataset based on a query; and retrieving data from the ranked data in response to the query.

Data retrieval using reinforced co-learning for semi-supervised ranking

Systems and methods for alleviating neck pain during media consumption. The methods and systems comprise determining, using control circuitry a current orientation of each of a user's head and a user device, determining an environmental parameter relating to the user's environment, determining a range of permitted orientations of the user's head based on the environmental parameter and the current orientations of the user's head and the user device, determining, using control circuitry, whether the orientation of the user's head is outside of the range of permitted orientations, and in response to determining that the orientation of the user's head is outside of a range of permitted orientations, causing, using control circuitry, a change in a mode of display of the user device.

Methods and system for alleviating neck pain during media consumption

Remote distribution of multiple neural network models to various client devices over a network can be implemented by identifying a native neural network and remotely converting the native neural network to a target neural network based on a given client device operating environment. The native neural network can be configured for execution using efficient parameters, and the target neural network can use less efficient but more precise parameters.

Remote distribution of neural networks

Systems, methods, and apparatuses are described for capturing panoramic images and positioning virtual objects on a device screen, using a device having a static camera and an adjustable camera. To generate a panoramic image, the device moves the field of view of the adjustable camera by moving a corresponding MEMS mirror. The device then captures a first image using the static camera, and a second image using the adjustable camera, and generates a panoramic image by combining the first and second images. To position a virtual object, the device captures a first image using the static camera, and determines that there are insufficient visual features in the first image for positioning. The device moves the field of view of the adjustable camera by moving the corresponding MEMS mirror, and captures a second image. Visual features from the second image are then used to position the virtual object.

System and methods for enhanced AR tracking via adaptive MEMS scanning mirrors

Sentiment classification can be implemented by an entity-level multimodal sentiment classification neural network. The neural network can include left, right, and target entity subnetworks. The neural network can further include an image network that generates representation data that is combined and weighted with data output by the left, right, and target entity subnetworks to output a sentiment classification for an entity included in a network post.

Multimodal sentiment classification

Growing community of inventors

Irvine, CA, United States of America

Ning Xu